(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "A COMPARISON OF PREFERENCE FOR TRAINING WITH AN EVALUATION OF COMBAT PERFORMANCE OF PILOTS"

SCHOOL OF AVIATION MEDICINE 

U.S. NAVAL AIR STATION . 
PENSACOLA FLORIDA 




A COMPARISON OP PREFERENCE FOR TRAINING 
WITH AN EVALUATION OF COMBAT PERFORMANCE. OF PILOTS 



Research Project X-718(Av-375-k) Report No. One 



1 July 1946 



NAVAL SCHOOL OF AVIATION MEDICINE 
U.S. NAVAL AIR TRAINING BASES 
PENSACQLA, FLORIDA 



RESEARCH REPORT 
SUBMITTED - 1 July 1946 



PROJECT NO. X-7l8(Av-375-k) 



REPORT NO. 



ONE 



TITLE t 



REPORT BY; 



A COMPARISON OF PREFERENCE FOR TRAINING 
WITH AN EVALUATION OF COMBAT PERFORMANCE 
OF PILOTS 




Trumbull" HIS) 7 USNR 



Lt, 



Ijg) Oscar Backs itrom, Jr f Srfs) 3 



USNR 



APPROVED : 



CAPTAIN 
MEDICAL 



SfTOUIS IVERSON, (MC) 



USN 



OFFICER IN CHARGE 



The opinions or conclusions contained in this report 
are those of the authors they are not to be construed as 
necessarily reflecting the views or the endorsement of the 
Navy Department,, Reference may be made to this report in 
the same way as to published articles noting, author , titl 
source, date s project number and report number <> 



SUMMARY 



A pilot' 3 desirability or un desirability in combat situa- 
tions, expressed by fellow pilots, is employed as a criterion 
of combat success for comparison with, preferences stated at 
Intermediate Training to determine to what extent such prefer- 
ences correlate with and/or predict combat performance. The 
preferences considered are (1) those for type of combat air- 
craft desired, and £2) those indicated on a check list of 150 
attitudinal items „ The relationship with individual items is 
determined by means of an item analysis while large factors or 
clusters of items are developed through multiple contingency 
procedures., Data are presented and discussed for the indi- 
vidual sources of preference to show the reliability and 
nature of each. In addition, multiple contingency provides 
an opportunity to study the relative contribution of each 
source of preference in one composite picture of preference vs 
combat success 



CONCLUSIONS 

1 More pilots who receive training for and duty in combat 
aircraft of high preference are rated as desired by 
fellow pilots than those whose assignments are in other 
preferences o The converse is true of men receiving low 
preferences „ 

2* Pilots who fly aircraft of 1st preference in combat after 
training for other, aircraft tend to fall in the LOW group „ 
Therefore, the consistency of training is a contributing 
factor* 

3» The pattern of preferences from most desired to least 
desired has a significance in addition to that of 1st 
preference in predicting combat success « 

4. In general, men of both CV and VP HIGH groups prefer any 
duty connected with flying to duty oia the ground • 

5e Men of both HIGH groups manifest a distaste for instructor 
duty of any kind, whether ground or flight » 

6. HIGH CV men, in particular, prefer active, social recrea- 
tion to inactive, solitary recreation. 

7* The factors discovered here measure relatively stable, 

unitary characteristics predisposing toward combat success, 

8. Items similar to those of the Aviation Preference Check 
List are capable of predicting combat success as established 
by the present criterion o 

9. A definite pattern of personality is emerging in the role 
of "The unwanted man" In combat which further study should 
synthesize* 



INTRODUCTION 



During the course of World War II, it was the policy of 
the Navy to give high priority to individual preference when 
assigning men to duty,, In aviation, this function was per- 
formed at the Intermediate stage of training by an Intermed- 
iate Training Selection Board which, by assigning a pilot to 
a certain type of specialized, training, largely predetermined 
the type of combat aircraft which he would fly. The extent 
to which preferences could be granted, however, was subject 
to various limitations? the needs of the fleet," and special 
qualifications required of the pilot for optimal operation of 
different types of aircraft. The latter included high alti- 
tude tolerance, height and weight specifications and response 
to psychological tests , While these devices were introduced, 
tried, modified or eliminated, the preference of the pilot to 
fly a certain type of aircraft remained the foremost criterion 
for placing him in training to that type of aircraft. 

It was not until the closing months of the war that data 
could be gathered for appraising any of the selection or 
placement techniques which has resulted in one man's flying a 
fighter type aircraft and another man's flying multi-engine 
aircraft. Beginning in November of 1944, a comprehensive 
study was undertaken in the combat areas of the Pacific to 
discriminate between the "good" combat pilots and the "poor" 
combat pilots. Prom the results of this study, judgments 
were obtained on over 2 S 000 pilots, on whom detailed analysis 
could be made from a multitude of factors contained in their 
past records. This present study is but a part of the major 
project of analysing this large body of data* It is concerned 
with comparing preference for aircraft type and selected items 
from an aviation preference check list with combat performance 
as established by the criterion to be discussedo 



SOURCES OF DATA AND PROCEDURE 

The data of this study were, obtained from: 

' (1) A master list of over 2,000 combat pilots, as rated 
by fellow pilots, prepared by the Aviation Psychology Branch 
of the Bureau of Medicine and Surgery from criterion data ob- 
tained in the Pacific area, 

(2) A pencil-and-paper test known as the Aviation Prefer- 
ence Check List (APCL) , taken by some of these pilots at the 
beginning of Intermediate training. 



- 1 - 



(3) Lists of preference for operational aircraft made 
out by some of these pilots at the beginning of Intermediate 
training . 

a» Preference lists: During the Indoctrination Week in 
Intermediate, the student was asked to fill out a preference 
sheet which listed six combat types of aircraft, VB2, VP, 
VO(VCS), VP, VTB, and VSB* (App« AI) After listing the six 
in order of preference, he indicated the degree of pleasure 
or displeasure which he would feel were he. to be assigned some 
other type, 

bo Aviation Preference Check List: (App All) This check 
list was one of two -questionnaires given during Indoctrina- 
tion Week for the selection of flight instructors o It con- 
sists of 150 items of two choices, one of which is to be 
checked as preferred over the other<, These items range from 
types of duty to magazines, activities, companions, et cetera. 
The check list was included in this study because it contains 
a number of items wherein the student must make a choice 
between one type of combat duty and another, as well as items 
in which a choice must be made between flight and non-flight 
duty,, It appeared to offer opportunities for a more definitive 
indication of preference than could be obtained from the pref- 
erence lists alone 

Co Combat criteria? In November of 1944, four H(S) psycholo- 
gists reported for duty in the Pacific area to interview com- 
bat pilots and obtain measures of combat proficiency for as 
large a group as possible o The procedure employed may be sum- 
marized as follows % Any group of pilots was gathered in a 
ready room or 1 other available area., The purpose of the pro- 
gram was outlined to them and stress was placed upon the fact 
that the responses would be secure and anonymous „ They were 
given prescribed forms (App AIII) to fill out after the fol- 
lowing introduction o 

For CV squadrons: "Assume that you have orders de- 
taching you at once from this Air Group and assign- 
ing you to another Air Group for a tour of combat 
duty Your orders specify that - regardless of your 
rank or your experience - you are to fly wing on 
someone in the new Air Group „ Write down on Line #1 - 
the name of any pilot, known to you personally, on 
whom you would be willing or pleased to fly wing in 
such an assignment a He may now be living or dead, 
may hold a rank above you or below you, and he may 
or may not be a member of your present Air Group 
The important thing is that he should be known to 



- 2 - 



you personally and that you would be willing 
or pleased to fly on him in combat." For VP 
squadrons: "Assume that you have been incapa- 
citated while on a combat mission and. that you 
must, turn the control of your plane completely 
over to your co-pilot . In this situation, of 
all the men you know in Naval Aviation, what 
man would you be best satisfied to have take 
over the controls to complete the mission and 
bring you back to your base? 

« The man you' choose may be of any rank and 

in any squadron- He may now be living or dead. 
The Important thing is that he should be known 
to you personally and that you would be satis- 
fied to have him take over completely the con- 
trol of the plane In which you are riding* ■ 

Additional information relative to the nominee's 
sqadron, etc., was obtained. During the first phase of the 
program, each respondent was asked to give his reasons for 
each nomination by free response. Tabulation of these rea- 
sons resulted in the devising of a check list of more common 
reasons for use in the later phase of the" program. In addi- 
tion to checking those reasons which applied to his nominee, 
the respondent had additional space for stating other reason 
not on the list which might be unique with that nominee. 
Respondents were requested to indicate further the three 
reasons which applied most specifically to their nominee. 
The same procedure was followed in nominating a second HIGH 
man. 

The second half of the procedure was similar to the 
above. Using the reverse of the form provided, responses 
were obtained for the following situations. 

For CV: "Again assume that you are transferred 
to a new squadron. This time your orders desig- 
nate you as a Section leader. You are authorized- 
to pick your own wingman from all the men known 
to you in Naval Aviation. In this situation, 
write on line 1 (here) the name of one man whom 
you very definitely do NOT want on your wing in 
combat. 

For VP: "Again assume that you are transferred 
to a new squadron. You are authorized to pick 
your co-pilot from among all the men you have 
known in Naval Aviation. Write on line 1 the 
name of one man, known to you personally, whom 
you do NOT want as your co-pilot on a combat 
mission. " 

- 3 - 



Reasons were indicated as before , the respondent 
being aided by a list of previously found "NOT 
wanted" traits „ The three most specific reasons 
were checkedo A second LOW man was listed with 
appropriate reasons in similar manner „ 

Prom this summary, it can be seen that from each respondent 
there were two nominees for a HIGH group and two for a LOW group ° 
It is evident that summation of these nominations would result 
in the formation of three groups ; HIGH where all nominations for 
the man were HIGH* LOW where all nominations for that man were 
LOW, and Indeterminate (-=) where the man concerned received nomi- 
nations for both categories 

Subjects % Because all data were not available for all pilots on 
the master list , several groups were employed In the various 
sections of this study to obtain the maximum N while observing 
requirements for st andardization The resultant groups and 

the ir composition follows 

Group 1, In comparisons showing the relationship of 
various items from the Aviation Preference Check List 
to combat success <, H - 502 □ This is the largest group 
of men from the master list for which Aviation Prefer- , 
ence Check Lists were available „ 

Group 2 o In comparisons correlating combat success 
with the correspondence of assignment and preference , 
N - ,321 o This is the largest group of men from the 
master list for which complete preference data were 
available „ 

Group 3 In comparisons combining both items from the 
preference lists and items from the Aviation Prefer- 
ence Check List , N - 199 This is the largest group 
of men for which all data were available All the 
members of this group are included in both groups (1) 
and (2) o 

The men of these groups were trained at Pensacola in classes 
entering Intermediate Training between June of 1943 and October 
of 1944 The average age of the. group at time of entering Inter- 
mediate was 21 years o The representative nature of the sample 
used might be shown in a number of ways Of greatest importance 
is the distribution of HIGH and LOW men. In the total group of 
2080 men ? 47o8$ were In the HIGH group „ In the test sample used 
for the-most comprehensive section of this study (N of 199) , 
48o6$ were so nominated,. The Indeterminate nominees were not 
employed in this study although some data are presented for this 



4 




group for the purpose of completing tables. The distribution 
of Group 3 in combat assignments, according to nominations, 
follows s 

Combat Aircraft *> 

Combat 



Rating 


F 


SB 


TB 


P 


HIGH 


48 


7 


13 


28 


LOW 


45 


_8 


18 


32 


- Total 


93 


15 


31 


60 



P designates VP (Fighters) 
SB designates VSB (Dive-bombers) 
TB designates VTB (Torpedo-bombers) 



VP P designates VPB (Patrol bombers, 

multi-engine ) 



Plan of analysis ; 

(1) The relation between combat success and cor- 
respondence of aircraft choice with assignment 
was investigated by means of standard contin- 
gency procedures, 

(2) The frequency of particular patterns of air- 
craft preference and the significance of cer- 
tain patterns in relation to combat success 

. were investigated by means of contingency pro- 
' . cedures* 

(3) The relation of choices on each item of the 
Aviation Preference Check List to combat suc- 
cess was determined by an item analysis in- 
volving a frequency count of responses to each 
item and the statistical significance of the 
resulting proportions of HIGH and LOW men. 



- 5 - 



cv 



(4) Maximal multiple correlations of combinations 
of items from all the above sources with com- 

. bat success were derived by means of multiple 
contingency o Factors predisposing toward com- 
bat success and combat failure were isolated 
and describedo 

(5) Because preliminary analysis indicated that 
important differences exist between CV men and 
VP men, the two classifications were treated 
separately in all comparisons o 



R ESULTS AND D1S CUSSI0H 

Preference t The relationship of preference to combat ratings 
can be shown by a study of the 321 men included in Group 2, 
Inasmuch as every pilot is not able to be assigned to fly the 
type of aircraft he desires , the nature of preferences and 
their background should be considered^ Of greatest importance 
is the fact that preference stated during Indoctrination Week 
at the Intermediate level of training have no foundation in 
experience <> With but few exceptions, the students' aviation 
experience had been confined to preliminary training in WTS 
and Primary with biplanes of the N2S type and monoplanes simi- 
lar to the popular Piper Cub type His acquaintance with the 
types of aircraft flown in combat and the nature of the duty 
involved was subject to bias from aviation magazines and the 
popular stories which glamorized to an extreme Some indica- 
tion of this influence is shown in trends of preference which 
reflect the impact of popularized exploits of squadrons or 
Air Groups o 4 particular type of aircraft might have a reputa- 
tion for containing innumerable "bugs 1 * jeopardizing the pilot « 
As illustrations, the PV and SGI both had their period of un- 
popular ity Or a particular aircraft might be associated with 
undesirable duty as was true with PBY's during the period of 
greatest replacement and development of Dumbo squadrons. 

To combat these misconceptions, pamphlets and lectures, 
were developed for use during Indoctrination Week prior to the 
time at which preferences were stated In addition, similar 
efforts were made at Primary bases where,, especially in the 
closing months of the war, definite lecture periods were estab- 
lished wherein returned combat men. informed the students of 
combat operations and greatly contributed to development of 
the proper perspective relative to preference o 

Among other factors influencing preferences are those re- 
lated to the social aspect or personality trait, e g , some 



= 6 - 



pilots duplicated a friend's list of preferences in an 
effort to maintain a previously established relationship* 
It is not intended that these examples produce the. conclu- 
sion that all preferenc-es had inadequate basis or reflected 
merely transient desire. Rather, they are mentioned so 
that one might constantly be aware of the factors which 
were influencing; aircraft choice. How closely these fac- 
tors were related to preference as well as combat profici- 
ency will be seen later in the personality and preference 
patterns found by means of multiple contingency,, 

Table 1 shows the preference frequencies of this group 
of 321 men. The popularity of certain aircraft and the un- 
popularity of others at the time this group arrived at 
Pensacola is clearly shown. The ideal situation would have- 
resulted in assignment of each man to training leading to 
duty in the aircraft of 1st preference. Due to previously, 
mentioned exigencies of war, availability of facilities, 
etc., some assignments had to be arbitrary. It was at this 
point that the other previously mentioned criteria for 
assignment were employed. The number of men receiving 1st 
preference at Intermediate, after the Intermediate Training 
Selection Board had considered these factors, was: 

Aircraft Type 

F SB TB J?_ _B_ VO Total 

No. stating as 

1st preference:- 121 22 15 65 95 2 321 

Assignments: 88 19 14 58 16 195 

$ assigned to . . , 

1st preference: 73$ 86$ 93# 89$ 17$ 0$' 61% 

Table 2 shows the ultimate assignment of all men in 
Intermediate and their final fleet duty. It is seen that 
65$ of the men actually saw combat in the aircraft for which 
they trained in Intermediate. A brief picture of this shift 
in assignment showing the number of 1st preferences, the 
assignments made at Intermediate and the billets ultimately 
filled from this group might be extracted from Tables 1 and 
2 as follows: 



- 7 - 



Aircraft Type 



1st Pref. 
Int. Assign. 
Fleet Duty 



p 


SB 


TB 


P 


121 


22 


15 


65 


103 


62 


65 


69 


150 


34 


48 


89 



B 

95 

20 



V0 
2 
2 



While the core of each operational squadron consists of 
men who- desired that duty, certain squadrons (TB in particular) 
show a wide discrepancy between the number of men who desire 
that aircraft and- the number of men which must be assigned to 
it for combat reasons „ This is clearly indicated by the dis- 
tribution shown in Table 3 where 49$ of the total group are 
shown to be flying 1st preference in combat. Further evidence 
is found in the analysis of Group 3 on page 5, which shows the 
HIGH and LOW nominations in terms of combat aircraft flown. 

The effect of consistent training in terms of preference 
Is shown by the three following analyses. First, the combat 
ratings of the group fortunate enough to receive training for 
and combat in the aircraft of 1st preference. B and V0 are 
eliminated due' to the small size of the groups involved.. 

Aircraft Type 



Combat 
Rating 


F 


SB 


TB 


P 


Total 


jL 


HIGH 


36 


2 


3 


32 


73 


52 




2 






1 


3 


2 


LOW 


32 


3 


4 


25 


: 64 


46 



This might be contrasted with an analysis of men who were 
assigned at Intermediate to meet quotas and whose combat duty 
was in the same type of aircraft as that assigned. This group 
is the same as the above, then, in completeness of training 
but they did not receive their 1st preference in the assign- 
ment • 



- 8 - 



Aircraft Typ e 



Combat 



Rat ing 


P 


SB 


TB 


F 


Total 




HIGH 


7 


6 


9 


3 


25 


40 








3 


2 


5 


8 


LOW 


6 


7 


15 


5 


33 


52 



Here we see that with all other conditions held constant, the 
influence of a man's receiving 1st preference is evident , One 
f urther group, that composed of men who fly aircraft of 1st 
preference in combat after receiving training in another type 
in the interim emphasizes the relation between preference and 
consistency in training □ 



Combat 
Rat ing 

HIGH 



LOW 



P 



10 



Aircraft Type 



SB 



TB 



Total 
6 
1 
11 



jl 

33 
6 
61 



This analysis of 1st preference per se presents the maxi- 
mum Influence of preference when granted and when not granted*, 
In addition, evidence is found establishing the Influence of 
consistent training once the preference has been granted,, The 
fact that a man received his first choice does not alone pre- 
dispose him to success "In combat but the fact alone that he re- 
ceives an assignment which was among his low choices does pre- 
dispose toward failure A complete picture of the part played 
by assignment of men to low preferences was precluded by the 
extent to which the Selection Board was successful in carrying 
out the intent of the overall policy of granting preference 
within limitations of quotas „ Thus, the men who received low 
preferences are so few In number that a definitive test of the 
effect of receiving them Is impossible., The indications are 
that if this sample had included more men in the low prefer- 
ence categories, the Importance of granting preference would 
have been readily discernible in the proportion of combat 
failures these categories would have . contained o 



- 9 - 



Preference patter ns g The relative preference that a pilot, 
e.g., a VP pilot s has for his combat assignment is shown not 
only in the rank order which he assigns F in his list of 
preferences (as first , second, third choice, etc), but also 
in the rank order assigned to other types and in the associa- 
tions within his list of F with other types For example , the 
three pilots whose orders of preference are reproduced below 
all placed F as 1st choice but the strength of their prefer- 
ence for F may be very disparate as might be the reasons for 
preferring F 



Pilot 


A 


- F 


SB 


TB 


VO 


B 


P 


Pilot 


B 


- F 


B 


P 


SB 


VO 


TB 


Pilot 


G 


- F 


P 


VO 


SB 


TB 


B 



The importance of such differences as appear in- these orders 
of preference may .best be realized if one imagines himself to 
be in the position of a member of an Intermediate Training 
Selection Board whose quota for VF can accomodate only half 
of those who list F as first choice „ Such a member evidently 
must consider the fact that the arrangement or pattern of 
preferences may have a significance in addition to that of 
the choices per se Moreover, it is evident that each choice ■ 
not only the first" and second choices - may be of significance 
in predicting combat success « It would be reasonable to hy» 
pothesize., for example, that' Pilot A, who shows a marked aver- 
sion to multi-engine aircraft (by listing B and P as fifth and 
sixth choices), might have an- additional predisposition for VF 
aircraft that Pilots B and C lack In other words , the low 
preference choices may have a positive Importance, as well as 
the negative importance which has been demonstrated 'In addi- 
tion, Pilot A demonstrates a pattern' of what will be termed 
"blocked" preferences; i o e , all single engine aircraft are 
in a sequence and both multi-engine aircraft are together 

An analysis of the patterns of preference was made in 
order to discover which groupings or clusters of choices pre- 
disposed to combat success and which predisposed to combat 
failure . For this analysis, the largest groups obtainable 
were used? i eo,- all men who had listed six choices on a 
preference list, (App<, I), were included whether or not other 
data were available for therm (It was possible to augment 
Group 2 for this analysis by including In the CV group men , 
whose fleet assignment was listed as "CV without specialty" 
or "Fighter-Bomber" „) The first step in this analysis was to 
tabulate, for the CV and VP combat groups separately, the 
number of times each aircraft type was listed In each posi- 



- 10 - 



tion by men rated HIGH and men rated LOW. These tabula- 
tions, which are shown in Tables 4 and 5, demonstrate the 
importance of choices other than 1st choice . Of the items 
which show potential value as predictors, about half tend 
to identify members of the HIGH group (positive predictors) 
and half tend to identify members of the EOW group (nega- 
tive predictors) o Following this tabulation process, all 
patterns were transcribed onto cards which were sorted manu- 
ally in order to discover which permutations and combinations 
of choices ; had potential value either as positive or negative 
predictors. Thirteen such combinations were identified,, 
They are shown in Table 6, together with percentage frequen- 
cies o Preliminary summations of these classes showed that 
while various permutations would aid prediction of success, 
especially for the VP group, additional variables were de- 
sirable. For this reason, no attempt was made to obtain 
maximum multiple coefficients with the pattern variables 
alone. It is worthwhile to draw special attention to the 
nature of one of these pattern variables: namely, "blocking" 
the choices. (See items 1 and 11) "Blocking" the aircraft 
types - that is, arranging them into generically similar 
groups - would appear to be the "normal", logical procedure. 
If one knew that he might not receive his first choice but 
was likely to receive one of his first three choices, he 
would "normally" make his second and third choices as similar 
to his first choice as possible; similarly, one would "nor- 
mally" group low choices as to type . Since this is so, devia- 
tions from "blocking" which are shown by a large number of 
individuals, such as Items 4, 5, 12, 13, and 14, are worthy 
of special notice, Inasmuch as they reflect particular modes 
of thinking and behavior which may be associated with combat 
success or failure . It is hazardous to impute motivations or 
reasons for these and other deviations but it seems safe to 
say that these patterns are,' distinctive and arise, if not from 
the same reasons in the case of every individual, at least 
from a common syndrome of reasons.* As such, they are in a 
sense measures of attitudes and/or traits as well as indica- 
tions of the preference or aversion. 



#A possible and plausible reason for the aversion to TB and 
VO may lie in the rumors which were once wide-spread among 
aviation cadets and air crewmen that TBF and 0S2U squadrons 
had abnormally high combat losses. In the course of Selec- 
tion Board, NAGS, and Interviewing duty, the authors many 
times heard both aircraft called flying coffins" by men who 
were unable to give a reason for the epithet* These and 
other aircraft also "developed" highly mythical defects of 
design and construction from time to time 



— 11 "* 



Aviation Pre ference Check List : ' The Aviation Preference Check' 
List, (App. All"), was devised as a test for Flight Instructor 
selection and was not designed to be a predictor -of combat suc- 
cess or accept ibility "Reasons for including it in this study 
have been mentioned elsewhere <, APCL answer sheets were avail- 
able for 502 of the combat men on the master listo Of the 150 
items which appear in this test, there were 34 which discrimi- 
nated between the HIGH and LOW groups at significant levels of 
confidence . These items with the preferences che.cked by the 
HIGH and LOW groups at the various levels of confidence are 
listed in Tables 7 and 8. As was mentioned previously, the 
GV men were separated from the VP men for analysis on the basis 
of early findings of dissimilarity. Table 7 presents the pref- 
erences of the CV group and Table 8 presents the preferences 
of the VP group o One caution in particular should be borne in 
mind In interpreting Tables 7 and 8; namely, that on only 5 out 
of the 34 items do a majority of the LOW men make a choice dif- 
ferent from the majority, of the HIGH men„ That is, as a group, 
the LOW men tend To make the same choices as the HIGH men, but 
the proportion of the LOW group which makes the "favorable 
choice Is significantly less than the proportion of HIGH men 
who do so. The typical LOW men , then, are distinguished from 
the typical HIGH men not by making all or even a large number 
of the LOW choices, but simply by making more of them than the 
HIGH men. Conversely, typical HIGH men make a number of LOW 
choices but fewer such choices than do the LOW men* Table 9 
illustrates this situation, using the 18 Items found to differ- 
entiate HIGH and LOW GV men of the smaller group (Group 3) « 
The score ds the number of items answered fav6raoly ; i.e., the 
number of HIGH choices The difference in the distribution of 
scores for the two groups Is highly reliable statistically, as. 
evidenced by a £ of ,00000 indicating not even 1 chance in 
100,000 that the difference could arise, from random fluctua- 
tion or "accident" o Moreover, it will be noted that it is 
possible to set various "cutting stores" which reject large 
proportions of the LOW men, but few of the HIGH men. For 
example, by requiring that men accepted for combat training 
have a score of 14 or more of the total of 18 Items, we would 
accept 89 *1$ of the HIGH men but only 44,6% of the LOW men 
and would reject 55,4$ of the low men, but only 10 o 3% of the 
HIGH men 

Unfortunately, only one item differentiated between the 
HIGH and LOW men of the smaller VP group, so a similar analysis 
for VP men was not possible As a matter of interest, this 
one item at a 1% level of confidence found the LOW men pre- 
ferring Ground School Instructor duty and' the HIGH men pre- 
ferring duty as a Meteorologist <> It was possible, however, 
to make one analysis of the VP men which produced rather inter- 



12 °- 



esting results . Scoring the VP men on the CV differentiating 
items, it was possible to establish that the VP men were more 
similar to the LOW CV men in their choices than they were the 
HIGH GV men. 

While one must proceed with caution in interpreting a 
limited number of items, certain consistencies are present and 
certain conclusions are warranted; 

1. In general, men of both CV and VP HIGH groups 
prefer any duty connected with flying to duty 
"on the ground o 

2. Men of both HIGH groups manifest a distaste for 
instructor duty of any kind, whether ground or 
flight. 

3. HIGH CV men in particular prefer active, social 
recreation to inactive, solitary recreation. 

4. A more nebulous distinction Is found in the 
greater frequency of men in the LOW CV group who 
prefer such items ass wear a dress uniform, 
military drill, associate with senior officers, 
arguments, tell others how to, referee a game, 
be leader in V- formation, give orders, and who, 
in general, are more inclined to accept ground 
duty with prestige value in preference to any 
flight billet „ There is a consistency which 
obtains in these items although, in toto, they 
seem to defy grouping under one heading or desig- 
nation as a personality type. 

The main contribution of this analysis is the demonstra- 
tion that items similar to those of the Aviation Preference 
Check List- may have considerable utility in predicting combat 
success. Certain items of this check list have been shown to 
have such value. Their use as a basis for another, extended, 
check list or questionnaire would serve a definite function 
in further development of selection tests. The authors are 
not unaware of the similarity between some of these items and 
those already incorporated in the Biographical Inventory which 
is used for initial selection in the flight training program 
where ability to succeed in the program was the criterion. 
Without a doubt, analysis under way on items of that test in 
relation to these same combat groups will produce certain items 
or indicators of personality patterns which, together with 
those gleaned from this check list, will be more than adequate 



- 13 



as a basis for further study . Prom such a composite, a mor.e 
definite classification of the personality type which seems 
to" exist here could be accomplished., Whether that type is an 
artifact of the. selection program in operation when these men 
were processed and the prevalence of such a type in society 
as a whole can be established in no other way. 



MU LTIPLE CONTINGENCY COMPARISONS AND ISOLATION OF FACTORS . 

The discussion has hitherto shown that the pilot's indi- 
cated choice of aircraft, the pattern of his preferences as 
to aircraft types, and his responses to various attitudinal 
items on the Aviation Preference Check List, all contribute 
to pre-estimation of the probability of his success in com- 
bat, No single variable which has been discussed Is alone 
predictive enough to be used economically in screening pilots 
for future combat duty While many of the variables discussed 
Improve prediction over chance, none of them by itself cor- 
rectly classifies a proportion of the group significantly 
large to enable selection authorities to accept only those 
men who pass the test item or to reject only those who fail 
it - the test item in this case being one expression of pref- 
erence, one particular pattern of preferences, one question 
on the Aviation Preference Check List, etc. Accordingly, it 
was sought to combine these variables in a manner which would 
maximize their predictive value . In general, there are two 
ways to combine a series of categorical (i.e., not numerical 
and non-continuous) 'variables such as these in order to pro- 
duce a composite test. The first, and simpler, method is to 
stipulate that a candidate must "pass" - that is, fall into 
the category which tends to be associated with the high 
criterion group - a certain number of the individual test 
items. This number is determined empirically by plotting the 
distribution of numbers of test items passed by both high and 
low criterion groups and choosing a dividing line or cutting 
score" which best separates the groups. This method was em- 
ployed in the comparison previously shown between HIGH and LOW 
groups on the significant items of the Aviation Preference 
Check List. It is important to note that this method does not 
stipulate which of the test items the candidate must pass,, but 
only the number . In other words, it does not take into account 
the fact that some test Items and, particularly, some combina- 
tions of test items are more predictive of success than others <, 
The other method Is to employ a technique of multiple associa- 
tion in order to select that combination of test Items- which 
will produce a maximum overall correlation with the criterion. 
Two standard techniques of multiple association, the techniques 



of multiple regression and multiple contingency, are available. 
The two techniques are similar in that each results in an over- 
all coefficient of correlation but differ in other respects. 
Each possesses peculiar advantages 

Because multiple contingency results in the formation of 
readily identifiable classes of individuals who pass or fail 
the same test items, it performs roughly the. same functions as 
factor analysis while simultaneously maximizing prediction. 
In other words, an overall coefficient of association is ob- 
tained and, at the same time, one may readily identify the under 
lying factors around which the various individual items cluster. 
In this way, one not only obtains a predictive battery, but also 
is enabled to arrive at some generalizations about HIGH men and 
LOW men. Accordingly, all statistically significant or promis- 
ing items for the CV and VP groups, respectively, were treated 
by multiple contingency, overall coefficients of correlation 
were obtained, and factors predisposing to combat success and 
failure were identified. 

Table 10 shows the contingency table which resulted from 
the application of multiple -contingency to the significant 
items of the CV group, with the contribution of each factor 
shown separately. The multiple coefficient of association is 
,57 which is high for selection batteries of this type and for 
a criterion which has possible subjective elements. The magni- 
tude of the coefficient is all the more remarkable because the 
CV group, being composed of VP, VSB, and VTB pilots, might be 
presumed to be heterogeneous in ways that would diminish such 
a coefficient, A more direct and graphic way of interpreting 
the table is to examine the frequencies in the four cells. It 
Is evident that by accepting only men who show the two positive 
factors, we would accept 50 out of 68 HIGH men, or 73,5% of the 
HIGH group and only 11 out of 72 LOW men, or 15.3% of the LOW 
group. Moreover, we would accept a group which would be com- . 
Sosed of 82^ (50/61) passers and 18^ (11/61) failures, instead 
of the 49$ (68/140) passers and 51% (72/140) failures which 
would be accepted if the two factors were not applied. Table 
11 shows similar contingency tables for the VP group, with 
results which are equally striking. Three contingency tables 
were prepared for this group: One (Table 11-A) which maximizes 

ttH.M. Johnson. Multiple contingency versus multiple correla- 
tion: an old time-saving way of handling multiple contingency. 
Am. J, Psychol ., 57, 1944, 49 - 63. 



- 15 - 



the ratio of accepted passers to accepted failures, but 
rejects a large proportion of potential passers along with 
the failures; and two (Tables 11-B and 11-C) which augment 
the ratio of rejected failures to rejected passers, but 
accept a large proportion of potential failures along with 
the passers o-sf By requiring all accepted VP candidates to 
show the characteristics which define the two positive 
classes in Table 11-A, we would accept IS out of 28 HIGH 
men, or 64 9 3% of the HIGH group, 'and only 4 out of 32 LOW 
men, or 12 „ 5% of the LOW group. The accepted group would 
be composed of 82% (18/22) passers and only 18$ (4/22) 
failures, and the rejected group would be composed of 74% 
(28/38) failures and 27% (10/37) passers. The re jected group 
contains 87% of the total number of LOW men, and 36% of the 
total number of HIGH men. Use of such a criterion would be 
based upon the potential supply of candidates inasmuch as 
where the supply is limited it would be impractical to re- 
ject a group containing so large a proportion of the poten- 
tial passers. Hence Tables 11-B and 11-C were also prepared. 
In these two tables, it is the rejected classes who are to 
be viewed as the factor-classes and the accepted classes as 
the remainders. In Table 11-B, it' may be seen that simply 
by rejecting all pilots who did not "block" their choices 
according to type of aircraft*«*, we would reject 56 „3% of 
the failures and only 21 e 4% of the passers. By "salvaging" 
the additional passers, however, we are forced to accept a 
group which contains 43*'8# of the failures, in contrast to 
the 12.5$ accepted by the previous table. The last table 
rejects only a sub-group of the group rejected by Table 11-Bo 
The group rejected did not "block" their preferences, and 
also expressed attitudes on the Aviation Preference Check 
List which differ from those expressed by the HIGH men. By 
limiting the rejection to this sub-group, we reject a gffioup 
composed of 100% failures. We accept a group which includes 



4* It is usually the case in such selection problems as these 
that several cutting points can be set, depending upon the 
primary purpose of the selection; i.e., whether the primary 
purpose is to reject the potential failures, or to accept 
potential passers. It is rare that one cutting point can be 
found which will accomplish both purposes as was true here 
for the CV group „ 



*-s:-This characteristic defines Factor-class C« 



- 16 - 



all the passers, but also includes 56.3$ of the failures, a 
proportion which may be impractically high for a training 
program to accept a 

For readers who are unfamiliar with correlation and con-, 
tingeney techniques, one note of caution should be inserted 
at this point; namely, that every selection battery must be 
subjected to successive trials on large samples before it can 
be regarded as stable and capable of producing the same results 
on samples other than the one from which it is derived,, It Is 
likely that if these batteries were applied to other groups of 
pilots the obtained correlation and obtained improvement in 
pre-classif ication would differ from those presented here. 
Consideration of the nature of the factors discovered, however, 
(see below) would indicate that these batteries measure rela- 
tively stable, unitary characteristics predisposing toward 
combat success which. a larger battery of "similar items chosen 
by comparisons on successive larger groups, "would measure with 
stability. In other words, while individual items of this 
battery to be discussed might not "hold up" and produce so 
good results on other groups, it is 'established a set of such 
items can be validated and cross-validated on other groups 
which would give results of useful predictive value », 

The Factors : Each of the factors named in Tables 12 and 13 
is defined by a set of characteristics shown by all members 
of the factor-class . In addition, each factor subsumes a set 
of characteristics which a large proportion, though not all, 
of the factor-class exhibit - i.e., characteristics which 
tend to "cluster" around a set of characteristics which de- 
fines the factor-class The nature of the individual factor 
was identified, and an appropriate label was selected for It 
by examination of all these characteristics. The importance 
of these factors, lies in the fact that they reveal the nature 
of the difference between the HIGH and LOW men. It is a truism 
that the items which go to make up a psychological instrument 
are imperfect measures and are subject to fluctuation In re- 
liability and predictive value. Presumably,, the nature of the 
entities they attempt to measure remains relatively unchanged. 
Knowledge of the nature of those entities makes possible in- 
creased accuracy of prediction through the addition of other 
tests which measure the same thing. These factors, then, may 
be viewed as a preliminary and tentative description of some 
of the characteristics which differentiate pilots who are 
"wanted" in combat from those who are "not wanted" . Each of 
the factor-classes shown in Tables 12 and 13 is thus composed 
of individuals who made the same responses to all of a certain 
set or constellation of items, and many of whom made the same 
responses to various other "clustering" Items. The fact that 



- 17 - 



members of any one class have a particular characteristic, 
then, does not preclude members of another class from also 
exhibiting it; distinction among the classes is primarily a 
question of the pattern of characteristics and the degree 
to which members of the various classes possess it. 

V P Factors : The primary positive factor among VP pilots 
(Table li'-A) is" labelled "Strong aversion to instructor duty" 
The reader may see for himself the indicated strength of this 
aversion by noting the responses of the members of this class 
to five items from the Aviation Preference Check List given 
in Table 12, It will be noted that of a variety of alterna- 
tives to various types of instructor duty, the members of 
this class are nearly unanimous in preferring those alterna- 
tives, whatever they are. The remaining pilots also express 
aversion to some types of instructor duty, but apparently it 
is not so markedo The particularly crucial items are Numbers 
1, 2 and 3, in which instructor duty is unanimously rejected 
by members of the factor-class, but is preferred by the major- 
ity of the members of the remaining classes . The second posi- 
tive class is labelled "Strong systematic preference for 
multi-engine, plus aversion to VO/VCS", All of the members 
of this class s 

(1) Listed B and P as their first and second choices, 

(2) Thus received either their first or second choice , 

(3) "Blocked" their choices, 

(4) Listed VO as their last choice (with TB as fifth 

choice • ) 

To a considerable extent, this class overlaps Factor-class A, 
Eleven of the 16 members of Factor-class A show characteris- 
tics {1}, (2), and (3), and in addition one out of the 16 may 
also be classed in Factor-class B The first three character- 
istics of choice pattern listed above are most positive when 
combined with the other characteristics, but that they are 
positive even when they are not "reinforced" by the other 
characteristics of Classes A and B may be seen by referring 
to Table 11~B The accepted group in this table is defined 
only by these characteristics The rejected group is defined 
by their failure to "block" their choices. It Is hazardous 
to interpret "blocking" as evidence of a particular faculty 
of logic, but this table is tentative evidence that the be3t 
VP pilots apparently make a choice not so specifically for VP 
as for multi-engine aircraft In general, as opposed to CV and 
VO, In other words, their preferences tend to be systematic. 



- 18 - 



As stated previously, the rejected class in Table 11-C 
is a sub-group of the rejected class in Table 11-B. This 
class is defined by two items i 

(1) Failure to "block" choices, 

(2) Stating preference for duty as Educational 
Officer rather than Meteorologist* 

In addition, the members of this class are inclined to accept 
tv/o other billets involving instruction which the members of 
the remainder class emphatically reject. It is not apparent 
what, if any, general tendency this class reflects, except the 
negation of the positive items of Factor-class A.. Table 11-C 
is offered simply as evidence that the pertinent items may be 
used to select either primarily for rejection or primarily for 
acceptance 

CV Factors ? Table 13 lists the make-up of two positive CV 
factor-classes, two negative CV factor-classes, and a negative 
remainder-class 

In the following discussion it will be important to bear 
in mind that the three negative classes are derived from the 
residual left by subtraction of the positive classes,, They are 
in other words, first defined as not exhibiting Factor A or 
Factor B c In addition, each has further defining characteris- 
tics, but these alone cannot be used to define the negative 
class | if they were, some cases in Factor-classes A and B would 
fall into the negative classes. 

Factor A is labelled in Table 13 "The wanted man". (After 
due deliberation, the designation "The -wanted man" was used as 
a standardization measure because this term had been employed 
In the restricted preliminary reports of the combat criterion 
study from the Aviation Psychology Branch in Washington The 
authors believe that the APCL suggestion of a definite person- 
ality type gains further support from the multiple contingency 
study and that further analysis will result in a more defini- 
tive designation.,) The set of characteristics which this 
appelation denotes perhaps will make easy a generalized picture 
of the "type" of man it represents, particularly to those who 
have done daily work in the flight training program. In genera 
this factor denotes? 

(1) A sort of non-systematic preference for CV air- 
craft or, at least, satisfaction with thenu CV 
preference Is evidenced by the high choice of 



™> 19 <■» 



P and SB, and by the low number of indi- 
viduals who made the multi-engine pair their 
first two choices. The. lack of the same 
sort of "system" that successful VP pilots 
apparently employed is evident in that none 
of these men placed three CV types as the 
first three choices; the only ones among them 
who "blocked" choices were 4 men who chose 
the multi-engine group * 

A susceptibility to fads or social trends in 
choices, as evidenced by the facts that: 

(a) Almost the entire group placed B among 
the first three choices. The interpreta- 
tion that this is a social trend, reflecting 
cadet-barracks "bull sessions", is based 
upon the fact that VP operations were re- 
ceiving widespread attention at the time this 
group made their choices, and the fact that 
the choice of B apparently is not indicative 
of a primary desire for multi-engine training. 
It occurs unassociated with P as often as 
associated with it, and in only six cases 
were the two listed as first and second. Also, 
in this group, the choice of P occurs only 
twice unassociated with B (i.e., without B 

in the first three choices as well) , whereas 
B is listed 23 times in the first three 
choices unaccompanied by P 

(b) TB does not appear among the first three 
choices . Reference to Table 1 will readily 
indicate that TB was highly popular among 
this group when choices were made. 

(c) TB and VO are listed as fifth and sixth 
choice by half of the group , Also, the mem- 
bers of this Factor-class who listed TB. and 
VO fifth and sixth constitute one half of the 
total group who did so even though the Factor- 
class is but 39% of that total sample* 

A preference for activities Involving action or 
popular socialized activities, in contrast to 
sedentary activities and work involving routine 
concentrat ion 



- 20 » 



CV Factor B in Table 13 is labelled "The wanted man with 
strong CV preference". Reference to the table will show where- 
in this label is appropriate. The two factors are distinguished 
entirely by items referring to the listed choices of aircraft. 
The two factors are virtually indistinguishable as to the per- 

L° r ^ ttit ^ de 1 - items from the Elation Preference Check 
Hi ; ,? 6 it ® m J . which dearly distinguishes the two classes is 
J? / a iJ- ur ! 2£ the members of Factor-class B to include B amonp 
their first three choices. Accepting Factor-classes A and B 
leaves a residual of 18 HIGH men and 60 LOT men. From this 
residual, two negative factors can be isolated. The largest 
cohesive negative class is defined by the failure of its mem- 
bers to include more than 1 CV type among their first three 
choices I he members of the class so defined, do not univer- 
sally Possess any other characteristic, in common, but a larae 
proportion of them possess one or more characteristics which 
make distinction of the class and interpretation of it rela- 
tively easy. Because of the tendency of the items which com- 
pose it and cluster around it, this factor is labelled "The 
misplaced multi-engine man or instructor" . It might particu- 
larly be noted that 74% of this class, which contains 24 LOW 

1 H^ J .f nd only 3 HIGH men » fl@w aircraft of their third, fourth 
fifth, or sixth preference in combat, as against 15 to 36$ of 
men of other classes who did so B 

Negative faster P ii strictly defined W ®&lf ene item. 

Motiuv i*f»i staled peferenee for teehnieal respensU 
eilifey rather than supervisory responsibility, ii §g ilisal» 
esieeiatid that it may fee usoi as well. Only ill mm in Shis 
fAeteii-eUii did not .state this preferonee. feeler 8 is a 
eiear antithesis fie peters A end 3. xt riiembln A In only 
one lte» Jp6) , and although it resembles 3 en the preference 
11 4 eleply differentiated from 3 ey items II! DO. 
end p*. lut while in opposition to the two positive lastlrs 
li elear, the nature of (fte f ao tov itself ii ionowhat elusive, 
Several el the responses listed ere oontradlstorr. end the* 
are alteiether difficult to harmonise SnlJii line foriuufii 
trm thelvoontradlotory natee, and from the nature if seme 
ef individual ehilsssi a type of persenaiity tulle dif« 
f event fm that HapUeit'in riSteri S ail It hott"a&H 
D subsumes is Ititt sen and i XX OH men. 

9he ?emaind»r»elass lift after suetrastlen if faetev- 
slaiMl o nnd 8 eensists ef li L&N wen rt »d I BIffl rteft, whn 
have in gammon a pattern if eheises oensisting ef S and twe 
0V types in the first to— • *ibiens ef preferenia, ike 
nUwn is readily diitin||Ui§liabie from I%rbo^6ialge8 n, 0, 
and D on fchs Basis ef profarense items, bub it in net Ils- 
I Lnguilhed from teeter- glass A by any single item. " 
memeers if this remainder-elesi have, in general, the ehar- 
aetorlstlos of faster-elan a, but not in a bhe earn* import iorni 
er the same patterns * F 

■ 31 m 



Comparison of CV and VP .Facto rs;; The reader probably has 
noted in the course of this discussion the difference in the 
type of items which distinguish HIGH from LOW VP pilots and 
those which distinguish HIGH from LOW CV pilots „ The most 
striking contrast which appears is the positive influence of 
"blocking" - that is, grouping choices according to type - In 
the VP group , and Its mainly negative influence in the CV 
group o The other outstanding difference is not a contrast, 
but leads one to Infer a possible difference in emphasis of 
attitude and outlook between successful VP pilots- and success- 
ful CV pilots | namely, the difference in the type of attitudi- 
nal items which have predictive value for each of the two 
groups o The Items successful in predicting VP success all 
pose a choice between instructor duty and some other duty-; 
whereas the Items successful in predicting CV success have In 
general to do with aspects of personality, tastes, and social 
habits o One serious limitation to the usefulness of such 
evidence and, perhaps to any inquiry based upon It, is the 
fact that these data were obtained upon a wartime population 
of Naval aviators who may hot have the same attitudes, tastes s 
and habits as succeeding peacetime populations „ Except for 
persisting prejudices or tenets,, attitudes are notoriously un- 
stable, both in the individual and' in the population at large 9 
and it is the more conservative procedure not to place great 
dependence upon attitudinal items until they have been thoroughly 
tested., -Fortunately,,- these comments do not apply to a large 
body of the biographical data and data from tests of aptitude 
and intelligence which has been accumulated during the war ; 
These tests do not depend so greatly upon attitude and may be 
expected to supply items which have both stability and predic- 
tive value o 

It remains to make a final estimation of the importance 
of choice. It has previously been noted that, while receiving 
his first preference or a high preference, does not predispose 
a man to inclusion in the HIGH group , receiving a low prefer- 
ence gives a slight predisposition toward Inclusion in the LOW 
group o It also has been" noted that the cases of assignment to- 
low preference are so few that its effect is not readily dis- 
cernible upon the total population., 

The multiple contingency comparisons presented above, how- 
ever, demonstrate unequivocally that the correspondence, between 
a man's choice and his fleet assignment is • highly correlated 
with success within sub-groups of the population,, That is, In 
combination with other characteristics predisposing toward com- 
bat success, satisfaction with obtaining aircraft of high pref- 
erence aids materially in placing a man in the HIGH group; in 



™ 22 ~ 



combination with other characteristics predisposing toward com 
bat failure, assignment to aircraft of low preference aids 
materially in placing a man in the LOW group It will be re- 
called, for example, that the fleet assignment of 74$ of the 
negative Pactor-class C of the CV group was 3rd, 4th, 5th, or 
6th preference o Other less tangible advantages with respect 
to morale and individual satisfaction add their weighty but 
the figures alone justify the conclusion that the practice of 
attempting to grant the expressed preference of the flight 
trainee pays off in later combat prof iciency<, 



■ ii ■ 



! 

. ■ , - . 

TABLE 1 

Frequen cy of preferences for Group 2 

■ 

■ 

Preference Aircraft Type 



Position 


F 




TB 


P 


_B__ 


VO 


Total 


1st choice 


121 


22 


15 


65 


95 


2 


320* 


2nd choice 


51 


69 


29 


65 


101 


6 


321 


3rd choice 


60 


76 


62 


28 


38 


57 


321 


4th choice 


48 


80 


59 


38 


57 


40 


322# 


5th choice 


24 


55 


79 


78 


18 


67 




6th choice 


18 


JL9 


77 


46 


12 


149 


321 


Total 


322# 


321 


321 


320* 


321 


321 






#0ne man 


listed 


F twice omitting 










TABLE 


2 








Distribution 


of Int« 


srmediate As 


9ignments VS 


Fleet 






Duty for 


Group 


2 






Intermediate 
Aao,l pjnmenfcs 


F 


SB 


Fleet 
TB 


Duty 
m P 


B 


VO 


Total 


P 


85 


12 


5 


1 






103 


SB 


38 


18 


5 


1 






62 


TB 


24 


4 


37 








65 


P 








69 






69 


3 


2 






18 






20 


VO 


— 1 




JL 








2 


Total 


150 


34 


48 


39 






321 



i 



TABLE 3 

Preference vs Combat Success 



Preference Position % due 

Criterion Fleet Choice to last 



Group 


Aircraft 


6th 


5th 


4th 


3rd 


2nd 


1st Total 4 


Choices 








4 


1 


6 


18 


42 


71 


15<,5 




r 












2 


2 




LOW 




2 


4 


6 


11 


10 


42 


75 


30 „7 


XlXLrfl 










5 


5 


2 


14 


50 oO 




DC 


















LOW 




1 


2 


4 


3 


7 


4 


21 


47„6 






2 




3 


7 


3 


3 


21 


71 4 




TB 






2 




1 




o 




LOW 






1 


1 


7 


11 


4 


24 


37 o 5 


HIGH 




1 


2 






10 


52 


45 


6„7 




P 










2 


2 


4 




LOW 




1 


_1 




J3 


11 


25 


42 


16,7 


Total 




7 


17 


20 


42 


78 


158 






i 




2o2 


5 o3 


6»2 


13 oO 


24 o 2 


49 al 







TABUS 4 



Frequency with which HIGH and LOW men in CV group 
listed each aircraft type in each position of pref erenceo 



Aircraft Type 



Pref . 


F 




SB 


TB 


F 




B 


VO 


Position 


Preq 


ST 

o % 

: * 


Freq, % 


Freq 


% 


Freq 


—?r 

( , % 


Freq, % 


Freq 




HIGH 


77 


51,7 


14 


9,4 


10 


6.7 


2 


1,3 


43 


28,9 


3 


2,0 


1st 


























LOW 


99 


54 ,1 


16 


8,7 


11 


6.0 


8 


4,4 


47 


25,7 


2 


1,1 


HIGH 


34 


22o8 


49 


32 a 9 


7 


4,7 


24 


16,1 


35 


23,5 





0.0 


2nd 


























LOW 


26 


14,2 


50 


27 3 


22 


12,0 


32 


17,5 


48 


26,2 


5 


2,7 


HIGH 


15 


10ol 


43 


28,9 


39 


26 , 2 


19 


12,8 


23 


15,4 


10 


6,7 


3rd 


























LOW 


27 


14,7 


50 


27.3 


45 


24.6 


27 


14,7 


23 


12,6 


11 


6,0 


HIGH 


11 


7,4 


33 


22 o X 


27 


18 o 1 


32 


21,5 


29 


19,5 


17 


11,4 


4th 


























LOW 


16 


8,7 


40 


21,8 


30 


16,4 


30 


16,4 


41 


22,4 


26 


14,2 


HIGH 


11 


7,4 


5 


3o4 


37 


24 ,8 


47 


31,5 


11 


7,4 


38 




5th 


























LOW 


11 


6,0 


22 


12,0 


37 


20,2 


58 


31,7 


13 


7,1 


42 


22,9 


HIGH 


1 


0,7 


5 


3o4 


29 


19,5 


25 


16,8 


8 


5.4 


81 


54,4 


6th 


























LOW 


4 


S3 o 2 


5 


2,7 


38 


20,8 


28 


15,3 


11 


6,0 


97 


53,0 



N - HIGH - 149 
IT - LOW - 183 



TABLE 5 

Frequency with which HIGH and LOW men in VP group 
listed each aircraft type in each position of prefere nce a 

Aircraft Type 



Prefo F SB TB P B VO 

Position Freq, % Freq o % Freq . % Freq a j Freq , % ^ e a °-l 

HIGH 2 4.3 1 2,1 0*0 54 72 c 4 10 21 3 OSO 
1st 

LOW 4 8*5 OoO 

HIGH 3 6 a 4 1 2.1 
2nd 

LOW 3 6„4 1 Sol 

HIGH 13 27 o7 1 2 1 
3rd 

LOW 16 34 .1 7 14 „9 

HIGH 17 36„2 17 36 e 2 
4th 

LOW 13 27 7 17 36 2 

HIGH 4 8o5 21 44o7 
5th 

LOW 3 6 4 17 36 2 

HIGH 8 17.0 6 12 8 
6th 

LOW 8 17.0 5 10,6 



N - HIGH - 47 
N - LOW - 47 






OoO 


27 


57„5 


16 


34ol 





0„0 





0.0 


10 


21.3 


33 


70 2 





OoO 





OoO 


14 


29o8 


25 


53 o 2 


4 


8o5 





6 





OoO 


4 


8o5 


29 


61 7 


1 


3 o 1 


3 


6 4 


5 


10 6 


15 


31o9 


9 


19 2 





OoO 





OoO 


4 


8o5 


7 


14 9 


1 


2ol 


1 


2 o X 


8 


17 o 


19 


40 4 


2 


4»3 





OoO 


1 


v i 


14 


29 8 


1 


2,1 





OoO 


12 




19 


40 „4 


1 


2d 





OoO 


13 


27 o 7 


25 


53 D 2 


1 


Sal 





OoO 


3 


17.0 



TABLE 6 

Common Patterns of Preference 



Percentage Showing 
Pattern 



Pattern 



VP GROUP 



lo "Blocked" according to similar 
type, as; B PVO F SB TB, 
TB F SB VO P B, 
VO F B SB TB P. 

2 First 2 choices include B and P. 

3o Last 2 choices include SB. 

4o First 3, but not first 2, choices 
include B and P. 



LOW 
(N-47) 



42 
76 
45 

15 



HIGH 
(13-47) 



83 
89 
57 



LOW 

(Egg) 



CV GROUP 

5. First 3 choices include 2 CV, 
1 non-CV. 



39 
78 
82 
26 
25 



6, Plrat 2 choices include F or SB* 

7. First 2 choices do not include TB* 
8 Last 2 choices include TB and VOo 
9. First 3 choices include B and P. 

10. First 3 choices inolude F, SB, and TB. 27 

11. "Blocked".* 31 

12 o First choice is B, but P is not 

among first 3. 8 

13. B among first 3 choices j choices 

not "blocked". 54 

14 o Last 2 ohoices inolude TB and VO} 

oholces not "blookad"* SI 



HIGH 

OPTO) 



46 
85 
88 
34 
16 
24 
27 

15 

61 
32 



externa 1 and 11 are idantioaU 



TABLE 7 



Paired Items from the Aviation Preference Check List 
which Differentiate HIGH and LOW CV Men 



HIGH 



LOW 



Level of A significantly larger A significantly larger 

Item Signif i- number of HIGH men than number of LOW men than 
Ho, cance LOW men made these choices HIGH men made these choices 



1 =01 Fighter pilot 

2 oOl Patrol boat pilot 

f[ ,07 Navy transport pilot 
4 07 Instrument flight instruc- 

tor 



Operations officer 
Meterologist 
Operations officer 
Athletic officer 



5 


. oOl- 


Navigator on patrol boat 


6 


oOl 


Navigator on patrol boat 


7 


o 05 


Navigation patrol bomber 


8 


,01 


Navy transport pilot 




01 


Patrol boat pilot 


10 


,01 


Operations officer 


IT 


,05 


Athletic officer 


12 


.01 


Go to movies 


IS 


oOl 


Visit a friend 


IT 


,01 


Sporting page 


Tb 


•OS 


Readers Digest 


IS" 


,05 


Play table tennis 


17 


,07 


Motorboating 


18 


,05 


Supervisory responsibility 


19 


,07 


Work involving few details 


20 


,07 


Work in one location 


21 


,07 


Reading 


22 


,05 


Be grounded for physical 






defect 


25 


,01 


Athletics 


24, 


,01 


Play a game 


2T 


05 


Follow orders 


26 


05 


Repair a motor 


T? 


,05 


Associate with junior 






officers 


28 


,07 


Wear fatigue clothes 


29' 


,07 


Be wing man in V formation 



Ground school instructor 
Student advisory officer 
Operations officer 

Primary flight Instructor 
Primary flight instructor 
Ground school instructor 
Ground school Instructor 

Read a book 
Write letters 
Comics 

Popular Mechanics 
Play cards 
Sailing 

Technical responsibility 
Work involving many details 
Change from place to place 
Arguments 

Be killed in crash 



Military drill 
Referee a game 
Give orders 
Tell others how to 
Associate with senior 

officers 
Wear a dress uniform 
Be leader in V formation 



The numbers of items which also differentiate the HIGH and LOW men of 
the smaller sample (Group 3) used In the multiple contingency compari 
sons are underlined© 



TABLE 8 



Paired Items from the Aviation Preference Check List 
which Differentiate HIGH and LOW VP Men 



HIGH LOW 

Level of A significantly larger A significantly larger 

Item Signifi- number of HIGH men than number of LOW men than 



No, 


cance 


LOW men made these choices. 


HIGH men made these choices 


I 


»01 


Navigator on patrol boat 


Student advisory officer 


2 


□ 01 


Navy transport pilot 


Primary flight instructor 


3 


o07 


Observation scout pilot 


Primary flight instructor 


4 


.05 


Meteorologist 


Ground school Instructor 


5 


.07 


Meteorologist 


Educational officer 


6 


.07 


Bomb as you saw fit 


Bomb under specific orders 


7 


o05 


Farm life 


City life 



TABLE 9 

Contingency table showing, for CV men, the correlation 
between combat success and a score based upon 



significa nt item s of the A F C L 

LOW HIGH 

Score Freq . % Freq~ |> Total 

18 0.0 2 2.9 2 

17 3 4.2 4 5.9 7 

Accepted 16 5 6.9 24 35.3 29 

15 13 18.1 21 30.9 34 

14 11 15.3 10 14.7 21 

13 11 15.3 5 7.3 16 

12 10 13.9 1 1.5 11 

11 13 18.1 1 1.5 14 

Rejected 10 2 2.5 2 

9 4 5.6 4 

Total 72 100.0 68 100.0 140 

St atistics : 
Chi-square = 40 „Q 



p (probability that the obtained chi-square arises from random 
fluctuation) = .00000 

Either 12, 13, or 14 may be used as a "cutting score", dividing 
the ACCEPTED from the REJECTED. 



TABliE 10 



Multiple contingency table showing for GV men the 
multiple correlation of various preference and attitudinal 
items, grouped into factors, with combat success. 

LOW HIGH 







Preq 


. _1 


Preq 


fo 


Total 




Factor-class A 


6 


8 ,33 


36 


52.92 


42 


ACCEPTED 


En pt of — c In a s R 


5 


6.95 


14 


20 , 58 


19 




Total accepted 


11 


15.28 


50 


73.50 


61 




Remainder- class 1 


28 


38.89 


4 


5.88 


32 


REJECTED 


Remainder-class 2 


13 


18.06 


5 


7.35 


18 




Remainder-class 3 


20 


27.78 


_9 


13,23 


29 




Total rejected 


61 


84.73 


18 


26.46 


79 


TOTAL 




72 


100.00 


68 


100.00 


K"140 



St atistics : 
Chi-square = 46,36 

p (probability that the obtained ohl-square arises from random 
fluctuation) = ,000000 

r (coefficient of correlation) « ,57 

j>;amma (number of misplaced individuals correctly reclassified 
by accepting only Factor classes A and B)a - 40 

r» (coefficient of oorrectlvlty)** » .§7 

•JiThe coefficients gamma and r' are ineludod primarily for readers 
familiar with contingency teehniquei. Gamma denotes "the number 
of Individuals who would most probably have been mal-classif led 
by a worthless test, or by chanoe, but who would have been properly 
reclassified by this tost", r' is the proportion of mal-olaeeif i- 
cation removed by this battery which would be present if a worth- 
less battery wore applied. These coefficients are derived and 
described in H#M« Johnson "A useful interpretation of Pearaonlan 
r in 2 X 2 contingency-tables," Am. J. P sychol., 57, 1944, 236- 
2"42, from which the definition of gamma !■ quoted* 



TABLE 11 



Multiple contingency table showing for VP men the 
multiple correlation of various preferences and attitudinal 
items, grouped into factors, with combat success. 



A ... L 

Grouped primarily to maximize proportion of accepted HIGH men* 

LOW HIGH 







Freq. 


_ 

R 


Freq 




Total 




Factor-class A 


3 


9.38 


13 


46.42 


16 


ACCEPTED 


Factor-class B 


1 


3 « 12 


_5 


17.86 


__6 




Total accepted 


4 


12 . 50 


18 


64.28 


22 


REJECTED 


Rema Inder 


28 


87.50 


10 


35.71 


38 




TOTAL 


32 


100 ,00 


28 


100.00 


60 



Statistics : 



Chi- square - 16,60 

p {probability that the obtained chi-square arises from random 
fluctuation) = .0002 

r (coefficient of correlation) « .53 

gamma (number of misplaced individuals correctly reclassified by 
accepting only Factor-classes A and B) - 15 

r» (coefficient of correctivity) ■ .31 

B 

Grouped primarily to maximize proportion of rejected LOVf men. 

LOW HIGH 
Freq . | Free,. g Total 

ACCEPTED Remainder 14 43.75 22 78.57 36 

REJECTED Factor-class C 13 56.25 __6 21.43 24 

TOTAL 32 100.00 28 100.00 60 

Stat istics x 

Chi- square * 7.54 

P {probability that the obtained chl-@quar© arise® from random 
fluctuation) = ,02 

r (ooeffloient of seprelatlon) « 9 35 

gamma (number of miipleoed Individual! eerrectly re olaa lifted by 
rejeefelng only Peotor-olaBi 0) ■ 10 

r' (ooeffloient of oorreotivlty) » ,34 



TABLE 11 (Continued) 



Another grouping primarily to maximize proportion of rejected 

LOW men* 



ACCEPTED Remainder 
REJECTED Factor-class D 
TOTAL 



LOW HIGH 

Freq . J Freq . f Total 

18 56,25 28 100.00 46 

14 45.75 __0 0.00 14 

32 100,00 28 100.00 60 



Statistics ; 
Chi-square - 15*97 

p (probability that the obtained chi-square arises from random 
fluctuation) = .0003 

r (coefficient of correlation) - .52 

gamma (number of misplaced individuals correctly reclassified 
by rejecting only Factor-class D) = 13 

r' (coefficient of correctlvlty) = ,42 



TABLE 12 

Composition of Fa ct or- and Reitointer- o lassesp VP men 



Response or 
Preference 
iirran gem e n t_ J.s : 



Percentag e of m en so respon ding 
Factor -class 



Remainder 
Class TOTAL 



Rather than : A (H-16) B (N-6 ) (N-38) (N-60) 



Factor A 


1 f Met eo ro lo gi st 


Educational 


100 


33 


86 


46 


( plus ) : 




Officer 






84 


45 


Strong Aver- 


8*Meteorologist 


Ground School 


100 


33 


sion to In- 




Instructor 






86 


43 


structor 


3*0bservation 


Primary Flight 


100 





Duty 


Soout Plane 


Instructor 








81 


4*Navy Transport 


Priuary Flight 


100 


50 


30 


Core Items 


Pilot 


Instructor 






79 


80 




5*Navigator on 


Student Advisory 


94 


63 




Patrol Boat 


Officer 











Factor B 
( plus ) : 
Strong 3ye 
tenia tic 
Preferenoe 
for Multi- 
engine , 
plus aver- 
sion to VO 

Cor* Itema 



6 P la 1st or 
End choice 
- 7 Both P and B 
are inoluded 
in first Z 
olioloea 
B Oho ices 

"bloeJcad" 
9 Both TB and VP 
are inoluded 
in laat 8 
ohoioaa 



Third - sixth 88 100 87 88 

Laat four 81 100 79 78 



Not "bloolced" 75 100 47 60 

First four 13 100 26 30 



*Theae items ara from the Aviation Preferenoe Ohaole Li at. 



TABLE 13 

Composition of Factor- a nd Rem ai nder-classes, CY Men 



Response or Prefer- 
ence Arrangement 1 3; 

COM ITEMS: 



Bather than: 



Factor Classes Remainder 
A B CD Class TOTAL 

(N-42) (N-20) (N-27) (N~24) (N-27) (N-140) 



Al,First three 


All CV 


100 


15 


100 





100 


72 


choices not all CV 
















A3 Fir3t two 


Include TB 


100 


65 


96 


55 


67 


77 


choices did not 
















include TB 
















AUfWork involving few 


Work involving 


100 


70 


15 


71 




64 


details 


many details 














A4$Visit a friend 


Write lettera 


100 


100 


70 


83 


59 


83 


Ajgoci/.'HP itbbj 
















A5 B Choloas not 


"Blocked" 


91 


35 


56 


29 


100 


67 


"blocked" 
















A6 a 3 among firat 


Last three 


88 





96 





100 


64 


three ohoioea 
















A7 F or 3B among 


Both F and SB among 


86 


100 


37 


100 


81 


79 


first two 


last four ohoioea 














ohoioea 
















ABoTirst three 


Other arrangement 


60 


15 








100 


40 


ohoioea Include 
















E CV, 1 non-OV 
















A9oLast two oho lots 


Other arrangement 


45 





62 





15 


26 


inolude TB and VO 
















Al0*3upervleory res- 


Taohnioal respon- 


69 


100 


S56 


4 


56 


57 


ponsibility 


sibility 
















iflgwfa m m m m * m ■« 















FACTOR A (plus) 
"The Wanted Kan" 



"Those iUuni are from the Aviation Preference Oheok List. 



TABLE 13 (Continued) 



Response or Prefer- 
ence A r rangement is : 

CORE ITEMS: 

BloFirst 3 choices 
include at least 
S CV type 

B2 F or SB among 
first 2 choices 

B3oB aioong last 
three choices 

B4?Visit a friend 

B5?Supervlsory res- 
ponsibility 

B6tGrounded for. 
physical defect 



Rather than: 



1 or more 



last four choices 
Among first three 

Write letters 
Technical respon- 
sibility 

Killed in a crash 



Factor Classes Remainder 
- B A C D Class TOTAL 

(W-20) (N-42) ( N-27 ) ( N-24 ) (N-27) (N-140) 



100 


60 





100 


100 


68 


100 


86 


37 


100 


81 


. 79 


loo 


12 


4 


100 







100 


100 


70 


83 


59 


83 


100 


69 


56 


4 


56 


5? 


100 


86 


96 


33 


67 


84 



ASSOCIATED ITEMS: 

B7 Assignment was 

first or second 

choice 
B8 4 First 3 choices 

were all CV type 
B9 Choices "blocked" 
BlOSWork involving few 

details 



3rd, 4th, 5th, or 
6th 

Other arrangement 

Hot "blocked" 

Work Involving many 

details ' 



85 


64 


26 


75 


81 


65 


85 








100 





38 


65 


9 


44 


71 





33 


70 


100 


15 


71 


44 


64 



FACTOR B (plus) 
"The Wanted Man with Strong CV Preferences" 



♦These items are from the Aviation Preference Check Liat« 



TABLE 13 (Continued) 



Response or Prefer- 
enoe Arrangement^ la : 

CORE ITEM: 

Cl.First 3 oho iocs 
included only 1 
or no CV type 



Rather than: 



2 or 3 CV types 



Factor Classes Remainder 
C A • B D Class TOTAL 

(N-27) (K^2) ( N-20 ) (K-24) (H-27) (N-I40) 



100 40 



32 



ASSOCIATED ITEMS: 

C2,B amoud first 3 

choices 
C3.Both B and P 

among first 3 

choices 
C4o Assignment was 

3rd, 4th, 5th, 

or 6th choice 
G5?Work involving 

many details 
C6tGround sohool 

instructor 



Last 3 

Other arrangement 
1st or End 



Work involving 
few details 
Athletic officer 



96 88 
93 36 



74 



36 






15 



85 30 

56 19 5 








100 




25 19 

29 56 
13 22 



64 

28 

35 

36 

23 



FACTOR C (minus) 
"The Misplaced Multi-engine Man or Instructor" 



♦These items are from the Aviation Preference Check List. 



TABLE 13 (Continued) 



Factor Classes Remainder 

Response or Prefer- D A B £ Class TOTAL 

enoe Arrangement ia: Rather than : ( N-34 ) ( N-42 ) ( N-20 ) ( N-27 ) ( N-27 ) (N-140) 

CORE ITEM: 





Other arrangement 100 


u 


DO 


U 


u 




were axX ov type 
















ASSOCIATED ITEMS: 
















DS. Choices "blocked" 


Not "blocked" 


71 


9 


65 


44 





33 


D3tTechnical respon- 


Supervisory respon- 


96 


31 





44 


44 


43 


sibility 


sibility 














D4?Athletic officer 


Ground school 


87 


81 


95 




78 


77 




instructor 














D5fWork involving 


Work involving many 


71 


100 


70 


I 5 


44 


64 


few details 


details 














DStKilled in a crash 


Grounded for physical 


6? 


14 





4 


33 


16 




defect 














D7?Primary flight 


Navy transport pilot 


63 


14 


10 


7 


22 


16 


instructor 
















D8?3ailing 


Motor boating 


50 


31 


30 


33 


26 


33 



FACTOR D (minus) 
"The Inconsistent Unwanted Kan" 



♦These items are from the Aviation Preference Check List. 



APP. AI 



NATO PENSACOLA 
SPECIALIZED TRAINING PREFERENCE SHEET 



NAME 
(Print]" 



(Last) 



(First) 



Twiddle ) 



(Date) 



FLIGHT CLASS 



RANK & CORPS 



AGE LAST BIRTHDAY 



HEIGHT 



WT. 



Preference for type of specialized training: 
(Specify VB2, VO/VCS, VP, VF, VTB, VS3) 



1st Choice 
2nd Choice 
3rd Choice 
4th Choice 
5th Choice 
6th Choice 



VB2 — Multi-Engine landplane 
VP — Multi-Engine seaplane 
VO/VCS — Observation or 

Scout Training 
VF — Fighters 
VTB — Torpedo 1 

Bombers Carrier 
VSB --Scout Training 

(Dive) 

Bombers 



If I were assigned VO/VCS I would be: If I were assigned VP 1 would b 

/ \ A* "I 4 .-.V. 4-^,3 ( \ rial 4 rrln -I- q rl 



) delighted 

) very pleased 

) pleased 

) no opinion 

} disappointed 

) very disappointed 

If I were assigned VSB I would be; 

) delighted 

) very pleased 

) pleased 

) no opinion 

) disappointed 

) very disappointed 

If I -were assigned VF I would be: 

) delighted 

) very pleased 

) pleased 

) no opinion 

) disappointed 

) very disappointed 



If 



If 



) delighted 

) very pleased 

) pleased 

) no opinion 

) disappointed 

) very disappointed 

I were assigned VTB I would be 

) delighted 

) very pleased 

) pleased 

) no opinion 

) disappointed 

) very disappointed 

I were assigned VB2 I would be 

) delighted 

) very pleased 

) pleased 

) no opinion 

) disappointed 

) very disappointed 



Have you had aviation experience previous to Naval flight training? 

Yes No If Yes, explain, giving length of service, type of plane, 

duties and responsibilities, etc. 



APPENDIX All 

AVIATION PREFERENCE CHECK LIST 



INSTOJCTICNS: 



Each of the following items consists of a pnir of activities, hobbies 
or flight assignments. In each case, imagine that you had to choose 
one of the two alternatives and then quickly choose. If A, black out 
the space marked A on the answer sheet; if 8, black out the space 
marked B. Don't omit any pair. Don't take time to consider the 
situation but indicate your first reaction. If in doubt, guess' 



1. A Fly o fast airplane 
B F1y n slow airplane 



2. a Deal frith things 
B Deal with people 

3. A Fly single engine planes 
B Fly multi-engined plane 

4. A Do a job by yourself 

B Supervise others on the job 



20. A People older than yourself 

B People younger than yourself 

21. A Contact flying 

B Instrument flying 

22. A Formation flying 
E Flying alone 

23. A "Air Trails" 
B "Air Facts" 



S. A Fly large heavy planet 
B Fly small light planes 



24. 



A Boxing 
B Tennis 



6. A Fly with others In the ship 
B Fly solo 

7. A Short flights 
B Long flights 

8. A Work involving few details 
B Work involving many details 

9. A Change from place to place . 
B Work in one location 



10. A Aerobatics 

B Instrument flight 



11. A Shore-based duty 

B Fleet duty 

II. A Fly low 

3 Fly high 

13. A Present a report In writing 
B Present * report verbally 

14. A Give order* 

B Follow orders 



25. A Poker 

B Bridge 

2fi. A Write letters 

B Visit a friend 

27. A " Don fighting" 
B Formation flying 

28. A Diving 

B Swimming 

29. A Wrestling 
B Golf 

30. A Hunting 
B Fishing 

31. A Handling horses 

B Operating a machine 



32. 



A Teaching adults 
B Teaching children 



33. A City life 
B Form life 



A Bead a book 
B Go to movies 



34. 



A Married 
B Single 



16. A Regular work hours 

B Irrefular work hours 

IT. A Play basketball or football 

B Coach basketball or football 



18. A Pilot a plane 

B Ride in a plane » crew member 



35. 



19. A Order people 

B Persuade people 

11-4529 



A Kadio 

B Navigation 



36. A Military drill 
B Athletics 

37. A Aerodynamics 
B Meteorology 

38. A Gunnery practice 

B Aircreft identification 



.iy. A Arguments 
!■! Reading 



59. A Play hockey 

B Play handball 



40. 



A Popular Mechanics 
B Keeders Digest 



60. 



A Wear a dress uniform 
B Wear fatigue clothes 



41. 



A Com i c s 

Sporting Page 



61. 



A Play cards 

B Play table tennis 



A A date 

B A " stag" party 



62. A Workout on parallel bars 
D Dive from high places 



43. 



A Trap shooting 
B Hunting 



63. A Referee a game 
Ft Play a fame 



44. 



A Sailing 

B Motor boating 



64. 



A Write a story or play 
B Direct a play 



45. A Reading a book 

B Seeding e magaiine 



65, A Technical responsibility 
B Supervisory responsibility 



46. 



A Hiking 
B Swimming 



66. 



A Race • horse 
B Train a horse 



47„ 



A Be only men «t "hen party" 
D Be one of many stags at dance 



67. A Study mathematics 
B Study history 



48. A Have student freeie on controls 
B Have ship catch fire in the air 



68. A Daylight bombing 
B Night bombing 



49. 



A Solo students yourself 

B Have check pilot solo your students 



69. 



A Bomb an eirbase 
B Bomb a city 



SO, A Teach students throughout primary 

B Specialize in one s tage of instruction 



70. A Shoot down an enemy plane 
B Strafe enemy infantry 



51. 



A Ride a motorcycle 
B Drive a car 



71. A Be grounded for physical defect 
B Be killed in a crash 



52. 



A Make a bomb run 
B Strafe a train 



72. A Be forced down at sea 

B Be captured by the enemy 



53. 



A Repair a motor 

B Tell others how to 



73. 



A Be bombed by enemy aircraft 
B Be strafed- by enemy aircraft 



54. 



A Dancing 

B Go to a show 



74. A Be skipper of a submarine 
B Be skipper of s destroyer 



55. 



A Associate with senior officers 
B Associate with junior officers 



75. 



A Make a parachute jump 
B Swim 3 miles to land 



56. 



A Be an ell round athlete 
B Be s champ in one sport 



76. 



A High altitude bombing 
B Dive bombing 



57. 



A Mountain climbing 
B Go! f 



77. 



A Rear gunner in a heavy bomber 
B Gunner in a torpedo plane 



58. A Western movies 
B Comedies 



78. A Be wing man in a V formation 
B" Be leader in a V formation 



7'*, A Hr- 1 11 ptovidc ii lighlri ■. tii*ii foj hoiiihcis ijij t 

P Pilol litlflllif*! imill'l itghlCI S* I CCII 



A 

li 



U|"*f St i "lis Of f S eel 
Allilr I H ( l r I 1 1 i- r 



sn. A Be Inst m the nvi'i iml 
1! Make- n In- 1 1 y- turn) hit'. 



Hi.. 



A "I II I (jciln Itnmlic ■ Pi I ill 

II Ohsrrvnl ion mwl Scout im: I'llm 



81 A lose your gunner 

K l<nse ymir navigator 



III). 



A Clroilitil School Insl riifini 
H Athlct , ,■ uf I i pi- 1 



K'i. 



A l<nmh as you sn» lit 

Ft Homli under sjicei fie onli-is 



102. A fighiet Pilot 

II Inst rumen I Flight Instructor 



8.1. 



A St i »r.- host i Ii- infantry 
U llomh s hostile fort 



tn.l. A- Otfsrrvnt inn Scouting Pilot 
II Primary Flight tnstrut'toi 



M. 



A Shoot dn»n an enemy in flames 
B Ilnmli n hostile destroyer 



104. A Supply <»( fieri 

II Kduc h t i mil I Ofliccr 



A Hi- for Led down at sen . . 

II Bp forced down on the desert 



Ids. A Torpedo Ilomher I'ilot 
II Fighter I'ilot 



86. A Bait out at 10,000 feet 

IS Lose your engine on takeoff 



I0fi. A Mrtcnrnlngi st 

II Student Advisory Offirer 



87. A Fight from formation 

H Fight in individual dogfights 



88. A fnitue in night comhat 
ft Fight in full daylight 



107. A CSrottnd School Instructor 

II lus t rumen I Flight Instructor 



irj£ 



A Fighter Pilot 

II Observation mid Scouting I'ilot 



89. 



A Be Air Attache in London 
H Re CO. of NKATI 



109* A I'rimnry Fl igM Instructor 
11 Student Advisory Officer 



10. 



A Navigator on Patrol Float 
n l'i4ot on Patrol Boat 



110. A Instrument Flight Instructor 
B Primary Flight Ins true tor 



91. 



A Dive Bonnier Pilot 
B Fighter Pilot 



111. A Navy transport pilol 
I) Opera! i on s n f f i ce r 



'S2. 



A Primary Flight Instructor 

Intermediate Flight Instructor 



112. A Instrument Flight Instructor 
11 Patrol Hunt Pilot 



A Primary Fl i ghl Instructor 
II Check Flight Instructor 



11.1. C A l'til ml I Ion I Pi let 

li Olisrrvatiou and Scouting Pilot 



A Instrument Flight Instructor 
I! Operational Flight Instructor 



114. A Kduca t i una I Officer 
II Meteorologist 



'IS. 



A Intermediate Flight Instructor 
U Oprrnt ionnt Flight Instructor 



115. A Operations Officer 

R Navigation Patrol ItmiiPrr 



A Signal Officer on a Carrier 
B Fighter Pi lot 



116. A Athletic Officer 

H Instrument Flight Instructor 



'17.. 



A At hie t i c Officer 
13 Put ml Boat Pilot 



<J8. 



A I'rimnry Flight Instructor 
It Torpedo Bomber Pilot - 



117. A Observation Scouting Pilot 
It Navy Transport Pi lot 



118. 



Primary Flight Instructor 
Metcorologi st 



119. A Fighter Pilot 

B Patrol Boat Pilot 



139. A Primary Flight Instructor 
B Supply Officer 



120. A Primary Flight Instructor 
B Ground School Instructor 



140. A Meteorologist 

B Petrol Boat Pilot 



121- A Sipply Officer 
B Meteorologist 



141. A Navigator on pBtrol Boat 
II Ground School Instructor 



122. A Navigation on PBtrol Boot 
B Primary Flight Instructor 



142. A Educational Officer 

B Instrument Flight Instructor 



123. A Navy Transport Pilot 
B Torpedo Bomber Pilot 



143. A Student Advisory Of ficer 
B Navigator on Patrol Boat 



124. A Torpedo Bomber Pilot 
B Dive Bomber Pilot 



JL44. A Ground School Instructor 
B Navy Transport Pilot 



12S. A Petrol Boat Pilot 

B Primary Flight Instructor 



145. A Instrument Flight Instructor 
B Operation! Officer 



126. A Fighter Pilot 

B Navigator on Patrol Beat 



146. A Athletic Officer 
B Meteorologist 



127. A Primary Flight Instructor 
B Navy Transport Pilot 



147. A Supply Officer 

B Instrument Flight Instructor 



128. A Primary Flight Instructor. 
B Athletic Officer 



148. A Navy Transport Pilot 

B Student Advisory Officer 



129. A Torpedo Bomber Pilot 
B Patrol Boat Pilot 



149. ' A Operations Officer 

B Ground School Instructor 



130. A Meteorologist 

B Instrument Flight Instructor 



150. A Ground School Instructor 
B Supply Officer 



131. A Navy Transport Pilot 
B Fighter Pilot 



132. A Patrol Boat Pilot 

B Observation and Scouting Pilot 



133. A Instrument Flight Instructor 
B Navigator on Patrol Boat 



134. A Torpedo Bomber Pilot 
B Operation* Officer 



lt$, A Navy Transport Pilot 

8 Instrument Flight Instructor 



1U. A OporetUne Officer 

• ftinty rtitht Inttrwattr 



l$J, A Or Wind School In» truster 
• Mtt«»r«!«tlet 



1M. A fighttf mot 

• Operetlone Offietr 



APP. AIII 



Todays Date 



Your name_^ ' Pile # 

(Last) (First) (Middle ) 

Rank . ' USNR Squadron 



#1 Name Rank Squadron 

(Last) (First) (Middle) 

The reasons I have circled below apply to this man: 

12 3 4 5 6 7 8 9 10 11 

12 13 14 15 16 17 18 19 20 21 22 



abode 

Other comments on this man: (Write below) 



#2 Name Rank Squadron 

(Last) (First) (Middle) 

The reasons 1 have circled below apply to this man: 

1 2 3 4 5 6 7 8 9 10 11 

12 13 14 15 16 17 18 19 20 21 22 



a b c d e 

Other comments on this man: (Write below)